Purchase Prediction Research Articles

Comparative Analysis of Ensemble Learning Techniques for Purchase Prediction in Digital Promotion through Social Network Advertising

Comparative Analysis of Ensemble Learning Techniques for Purchase Prediction in Digital Promotion through Social Network Advertising

APPLICATION OF RFM MODEL ON CUSTOMER SEGMENTATION IN DIGITAL MARKETING

APPLICATION OF RFM MODEL ON CUSTOMER SEGMENTATION IN DIGITAL MARKETING

General Commerce Intelligence: Glocally Federated NLP-Based Engine for Privacy-Preserving and Sustainable Personalized Services of Multi-Merchants

One of the most crucial capabilities in the commercial sector is a personalized prediction of a customer's next purchase. We present a novel method of creating a commerce intelligence engine that caters to multiple merchants intended for the UB Platform, managed by e-payment company Harex InfoTech. To cultivate this intelligence, we utilized payment receipt data and created a Natural Language Processing (NLP)-based commerce model using a Transformer to accommodate multinational and merchant trade. Our model, called General Commerce Intelligence (GCI), provides a range of services for merchants, including product recommendations, product brainstorming, product bundling, event promotions, collaborative marketing, target marketing, and demand fore-casting etc. To bolster user privacy and foster sustainable business collaboration, especially among micro-, small-, and medium-sized enterprises (MSMEs), the GCI model was trained through federated learning, especially with glocalization. This study delves into the structure, development, and assessment of GCI, showcasing its transformative capacity to implement User Centric AI and re-shape the global commerce landscape to benefit MSMEs.

From data to design: AI practice in design decisions

Abstract Facing the increasingly fierce market competition, enterprises need to make correct marketing decisions. Based on the decision support system, cloud server, and SOA framework, the article establishes an enterprise intelligent marketing decision-making system by combining the TCP/IP communication protocol and designing a data warehouse to store the marketing decision-making data. The Apriori algorithm in association rules is used to generate marketing rules to assist marketing decision-making, the marketing data is classified using the random forest algorithm, and the sparrow search algorithm optimizes the parameters of the random forest model. In practice, simulation experiments were designed to analyze the performance of the random forest model and the marketing decision-making system, and they were applied to the marketing decision-making example of the e-commerce platform. It was found that the AUC value obtained when the number of binary trees of the random forest model was 75 was 0.836, and the number of conflicts per second of the system was only 0.41 when the number of concurrent users was 5*104. The use of association rules resulted in eight marketing decisions, with the highest probability of user purchase prediction reaching 98.54%. Based on the intelligent marketing decision system, it can help decision-makers better analyze market changes and provide decision support for enterprises to develop scientific and reasonable marketing strategies.

Advancing e-commerce user purchase prediction: Integration of time-series attention with event-based timestamp encoding and Graph Neural Network-Enhanced user profiling.

In recent years, the global e-commerce landscape has witnessed rapid growth, with sales reaching a new peak in the past year and expected to rise further in the coming years. Amid this e-commerce boom, accurately predicting user purchase behavior has become crucial for commercial success. We introduce a novel framework integrating three innovative approaches to enhance the prediction model's effectiveness. First, we integrate an event-based timestamp encoding within a time-series attention model, effectively capturing the dynamic and temporal aspects of user behavior. This aspect is often neglected in traditional user purchase prediction methods, leading to suboptimal accuracy. Second, we incorporate Graph Neural Networks (GNNs) to analyze user behavior. By modeling users and their actions as nodes and edges within a graph structure, we capture complex relationships and patterns in user behavior more effectively than current models, offering a nuanced and comprehensive analysis. Lastly, our framework transcends traditional learning strategies by implementing advanced meta-learning techniques. This enables the model to autonomously adjust learning parameters, including the learning rate, in response to new and evolving data environments, thereby significantly enhancing its adaptability and learning efficiency. Through extensive experiments on diverse real-world e-commerce datasets, our model demonstrates superior performance, particularly in accuracy and adaptability in large-scale data scenarios. This study not only overcomes the existing challenges in analyzing e-commerce user behavior but also sets a foundation for future exploration in this dynamic field. We believe our contributions provide significant insights and tools for e-commerce platforms to better understand and cater to their users, ultimately driving sales and improving user experiences.

Optimizing E-Commerce Profits: A Comprehensive Machine Learning Framework for Dynamic Pricing and Predicting Online Purchases

In the online realm, pricing transparency is crucial in influencing consumer decisions and driving online purchases. While dynamic pricing is not a novel concept and is widely employed to boost sales and profit margins, its significance for online retailers is substantial. The current study is an outcome of an ongoing project that aims to construct a comprehensive framework and deploy effective techniques, leveraging robust machine learning algorithms. The objective is to optimize the pricing strategy on e-commerce platforms, emphasizing the importance of selecting the right purchase price rather than merely offering the cheapest option. Although the study primarily targets inventory-led e-commerce companies, the model's applicability can be extended to online marketplaces that operate without maintaining inventories. The study endeavors to forecast purchase decisions based on adaptive or dynamic pricing strategies for individual products by integrating statistical and machine learning models. Various data sources capturing visit attributes, visitor details, purchase history, web data, and contextual insights form the robust foundation for this framework. Notably, the study specifically emphasizes predicting purchases within customer segments rather than focusing on individual buyers. The logical progression of this research involves the personalization of adaptive pricing and purchase prediction, with future extensions planned once the outcomes of the current study are presented. The solution landscape for this study encompasses web mining, big data technologies, and the implementation of machine learning algorithms.

Business Purchase Prediction Based on XAI and LSTM Neural Networks

The black-box nature of neural networks is an obstacle to the adoption of systems based on them, mainly due to a lack of understanding and trust by end users. Providing explanations of the model’s predictions should increase trust in the system and make peculiar decisions easier to examine. In this paper, an architecture of a machine learning time series prediction system for business purchase prediction based on neural networks and enhanced with Explainable artificial intelligence (XAI) techniques is proposed. The architecture is implemented on an example of a system for predicting the following purchases for time series using Long short-term memory (LSTM) neural networks and Shapley additive explanations (SHAP) values. The developed system was evaluated with three different LSTM neural networks for predicting the next purchase day, with the most complex network producing the best results across all metrics. Explanations generated by the XAI module are provided with the prediction results to the user to allow him to understand the system’s decisions. Another benefit of the XAI module is the possibility to experiment with different prediction models and compare input feature effects.

Predicting online customer purchase: The integration of customer characteristics and browsing patterns

Despite the relentless growth of online retail, e-commerce platforms still suffer from a low purchase conversion rate. Researchers and practitioners have attempted to understand customer purchase behavior, but it remains elusive due to customers' heterogeneous and complex decision-making processes. This study addresses a gap in existing research by combining two types of factors affecting purchase behavior: past customer characteristics and current website browsing behavior. This study employs recency, frequency, and monetary value (RFM) to extract variables of customer characteristics and utilizes graph metrics to comprehensively measure browsing patterns. Based on these variables, this study conducts predictive analysis for purchase behavior and clustering analysis for session-level customer segmentation. Our findings reveal that integrating customer characteristics and browsing patterns significantly enhances purchase prediction and proposes a novel customer segmentation. This study not only provides theoretical and methodological contributions to decision support and e-commerce literature but also offers practical insights for real-time personalized marketing strategies.

TEE: Real-Time Purchase Prediction Using Time Extended Embeddings for Representing Customer Behavior

Real-time customer purchase prediction tries to predict which products a customer will buy next. Depending on the approach used, this involves using data such as the customer’s past purchases, his or her search queries, the time spent on a product page, the customer’s age and gender, and other demographic information. These predictions are then used to generate personalized recommendations and offers for the customer. A variety of approaches already exist for real-time customer purchase prediction. However, these typically require expertise to create customer representations. Recently, embedding-based approaches have shown that customer representations can be effectively learned. In this regard, however, the current state-of-the-art does not consider activity time. In this work, we propose an extended embedding approach to represent the customer behavior of a session for both known and unknown customers by including the activity time. We train a long short-term memory with our representation. We show with empirical experiments on three different real-world datasets that encoding activity time into the embedding increases the performance of the prediction and outperforms the current approaches used.

Single and Multiple Separate LSTM Neural Networks for Multiple Output Feature Purchase Prediction

Data concerning product sales are a popular topic in time series forecasting due to their multidimensionality and wide presence in many businesses. This paper describes the research in predicting the timing and product category of the next purchase based on historical customer transaction data. Given that the dataset was acquired from a vendor of medical drugs and devices, the generic product identifier (GPI) classification system was incorporated in assigning product categories. The models built are based on recurrent neural networks (RNN) and long short-term memory (LSTM) neural networks with different input and output features, and training datasets. Experiments with various datasets were conducted and optimal network structures and types for predicting both product category and next purchase day were identified. The key contribution of this research is the process of data transformation from its original purchase transaction format into a time series of input features for next purchase prediction. With this approach, it is possible to implement a dedicated personalized marketing system for a vendor.

A gradient boosting classifier for purchase intention prediction of online shoppers

Early purchase prediction plays a vital role for an e-commerce website. It enables e-shoppers to enlist consumers for product suggestions, offer discount and for many other interventions. Several work has already been done using session log for analyzing customer behavior whether he performs a purchase on the product or not. In most cases, it is difficult to find out and make a list of customers and offer them discount when their session ends. In this paper, we propose a customer's purchase intention prediction model where e-shoppers can detect customer's purpose earlier. First, we apply feature selection technique to select best features. Then the extracted features are fed to train supervised learning models. Several classifiers like support vector machine (SVM), random forest (RF), multilayer perceptron (MLP), decision tree (DT), and XGBoost classifiers have been applied along with oversampling method for balancing the dataset. The experiments were performed on a standard benchmark dataset. Experimental results show that XGBoost classifier with feature selection techniques and oversampling method has the significantly higher area under ROC curve (auROC) score and are under precision-recall curve (auPR) score which are 0.937 and 0.754 respectively. On the other hand accuracy achieved by XGBoost and Decision tree are significantly improved and they are 90.65% and 90.54% respectively. Overall performance of the gradient boosting method is significantly improved compared to other classifiers and state-of-the-art methods. In addition to this, a method for explainable analysis on the problem was outlined.

Modeling of user profiles of financial products and comparison of purchase prediction models: Based on machine learning

In this paperwork, the Apriori algorithm can be used to describe the association between data in a database. This paper will use the demographic and financial information of the users in the data set to analyze the user profile as the basic information. This work will fit the model by optimizing the Apriori algorithm, generating a combination of two or more associated variables, and then using the decision tree to learn and output the final model. The main contribution of this paper is to explore the optimization of Apriori and the ef-fect of Apriori's combined model with the decision tree in predicting user buying behavior. Therefore, the conclusion can be drawn that in the sale of deposit products, if the custom-er has the above characteristics, such as no credit fault, no loan, housing loan, with a col-lege degree, marital status is single, and act as an administrator. The people mentioned above may be more interested in the product. In other words, it is more likely to sell suc-cessfully. In future work, in addition to deposit products, more customer buying infor-mation is highly needed to extend predictable results to broader anger.

Prediksi Pembelian Barang Pada Distributor Lampu Menggunakan Metode Apriori pada PT. XYZ

Apriori is an algorithm that is widely used to determine the pattern of relationships between products that are often bought together in a store. This Apriori algorithm will be suitable to be applied in the field of determining strategy or promotion. PT. XYZ is one of the Lighting Distributor Companies. One of the problems of the company is the imbalance of stock in the warehouse. So, the purpose of this research is to obtain product purchase predictions from suppliers to maintain stock balance with product solds to customers. The object of this research is the application of apriori algorithm. This research data is in the form of sales transaction data at PT. XYZ. Data analysis using RapidMiner Software by determining the valure of support and confidence. The analysis shows that there are 22 Association Rules with a minimum support 20% and a minimum confidence 50%. This information is excpected to help the company in developing a product purchasing strategy from the supplier.

Prediction of Car Purchase based on User Demands using Supervised Machine Learning

One of the key sectors of the national economy is the auto industry. Cars are becoming more and more common as a form of private transportation. When a buyer wants to purchase the ideal vehicle, particularly a car, an evaluation is necessary. Because it is an expensive vehicle, there are a lot of conditions and elements to consider before buying a new one, including price, headlamp, cylinder volume, and spare parts. Therefore, it is crucial for the consumer to choose a purchase that can meet all of the criteria before making any other decisions. In our research, we therefore suggest various well-known methods to improve accuracy for a car purchase. These algorithms were used on our dataset, which consists of 50 data. With a prediction accuracy of 86.7%, Support Vector Machine (SVM) produces the best result of the bunch. In this study, we also present comparison findings for all data samples using various methods for precision, recall, and F1 score.

Effect of the differential discount value on the purchase prediction: an approach from associative learning

The purpose of this study was to explore the effect of changing the context in which a product with different discount values has been associated with purchase prediction. Participants were trained in a learning task where they had to predict the probability of a result (purchase of a product) based on different images (products), presented in different contexts (stores). In store A, product X associated with different discount values (50%, 30%, 0%) and the product Z associated with a constant discount (30%) were presented. In store B, product Y was presented without discount (0%). Control products without discount were presented in each context (F1 and F2). The results indicate that the differential discount value is context dependent, which means that the probability of purchase is greater when the product with a variable discount is presented in the same store where the association with multiple discounts was previously learned. This effect was less with the product with constant discount and was not observed in the product without discount. The implications of these findings are discussed considering the effectiveness of discount promotional strategies in stores.

Influential Factor Analysis and Prediction on Initial Metro Network Ridership in Xi’an, China

To satisfy the adaptability of forecasting the short-term and abrupt volume of the initial metro network, we build the multiple enter linear regression (MELR) model to explore the determinants and forecast the intensity during the twice expansion of the initial metro network in Xi’an. We further compare the prediction of the metro transport capacity between the MELR models with exponential smoothing and autoregressive integrated moving average (ARIMA) models. Results show that the passenger intensity significantly fluctuates with the months and days, and MELR model is more adapted for the short-term prediction of the abrupt volume than the ARIMA model during the new metro line opening and the old line expands, which avoids the drawback of time series models that need a huge database. This study provides a guide for the prediction of initial metro network volume and accurate purchase of the rail vehicles during the metro planning and expends stages.

An analyses of the effect of using contextual and loyalty features on early purchase prediction of shoppers in e-commerce domain

Online sales have been growing rapidly in recent years. With the growing competition, online retailers have been keen to increase the effectiveness of their e-commerce platforms by providing a more personalised experience and increasing the ”conversion rate” (i.e. the proportion of visits ending in sales). The early identification of those customers who are likely to buy items could significantly improve the ”conversion rate”. In this paper, we present a novel framework of early purchase prediction in online sessions for registered and unregistered consumers as soon as they land on an e-commerce platform. Also, the paper provides extensive analysis of the performance of different data mining models using the proposed framework. Computational experiments on real-world datasets show that the proposed framework produces good results when appropriate session features are selected in the data mining model training stage, even when no products are browsed during the session. Contextual features without navigational data in the sessions can be used for early detection. When users arrive at the e-commerce platform, before any item interaction, we are able to predict which sessions will result in purchases early, with a high accuracy of 90.2 %. When we combine users’ past number of visits and purchase data, the performance has an even a higher accuracy of 95.6 %. The findings in this paper provide an understanding of context features and users’ loyalty related features that can help online shops’ marketing strategies as well as delivering a better user experience through personalised offers and discounts based on users’ early purchase predictions.

Data Mining Assisted Purchase Prediction

With the revolution from physical businesses to shopping online, predicting client behavior in e-commerce is becoming increasingly important. It can increase customer satisfaction and sales, resulting in higher conversion, by enabling a more individualized shopping process. Today, most users want to save their time using and they prefer to shop using the platform provided for e-commerce. Millions of transaction records are available in the databases of such websites using which, a customer shops something. Using the transaction to find something can be helpful for the organization or merchants. Using the available databases or datasets, to find some useful pattern can increase the business, to check out the customer satisfaction level, to check the customer behavior about the product, etc. Some of the useful information can be to find out which item will be purchased by the customer in the next visit, or which new items can be purchased by the customer in the next visit. Using this information, an organization or Shops can control the quantity and increase the maximum purchased items, improving the quality of products for the customers. For this purpose, we use supervised learning techniques for prediction. Because most of the data which we will use will be labeled. Many researchers used supervised methods but some of the researchers also used unsupervised methods too. We created a supervised model for predicting the basket items. Due to the large dataset, it was very difficult to extract the features and it takes a lot of time. We have performed feature engineering, to choose the best ones. After the training model, our model shows better performance than the previous results.

A Neural Network Extension for Solving the Pareto/Negative Binomial Distribution Model

This research proposes a new estimation scheme to solve the estimation burden of the Pareto/NBD model and to release its power in out-of-sample prediction and then builds a neural network-based model (NNA-based model) for parameter estimation of the Pareto/NBD model, by designing a loss function to include the likelihood of the Pareto/NBD model and the mean absolute error. Following empirical analysis with a real dataset and simulation datasets, the main results are as follows. (1) The NNA-based model encounters a relatively skewed prediction in inactive prediction and focuses more on inactive prediction, while the heuristic method has a balanced prediction. (2) The Pareto/NBD model has better repeat purchase prediction compared to that under the mean absolute error, but the NNA-based model has a better fitting in the repeat purchase distribution. (3) The distribution of estimated parameters between these two models is quite different, but is able to realize the same prediction purpose. The NNA-based model also has greater application opportunities in the commercial environment because the learned estimator is applicable to different transaction timing patterns. These findings clarify that the NNA-based model solves the estimation burden and releases the predictive power of the Pareto/NBD model.

Sales Prediction and Product Recommendation Model Through User Behavior Analytics

The COVID-19 has brought us unprecedented difficulties and thousands of companies have closed down. The general public has responded to call of the government to stay at home. Offline retail stores have been severely affected. Therefore, in order to transform a traditional offline sales model to the B2C model and to improve the shopping experience, this study aims to utilize historical sales data for exploring, building sales prediction and recommendation models. A novel data science life-cycle and process model with Recency, Frequency, and Monetary (RFM) analysis method with the combination of various analytics algorithms are utilized in this study for sales prediction and product recommendation through user behavior analytics. RFM analysis method is utilized for segmenting customer levels in the company to identify the importance of each level. For the purchase prediction model, XGBoost and Random Forest machine learning algorithms are used to build prediction models and 5-fold Cross-Validation method is utilized to evaluate their. For the product recommendation model, the association rules theory and Apriori algorithm are used to complete basket analysis and recommend products according to the outcomes. Moreover, some suggestions are proposed for the marketing department according to the outcomes. Overall, the XGBoost model achieved better performance and better accuracy with F1-score around 0.789. The proposed recommendation model provides good recommendation results and sales combinations for improving sales and market responsiveness. Furthermore, it recommend specific products to new customers. This study offered a very practical and useful business transformation case that assists companies in similar situations to transform their business models.